Dekaplane is not positioned as a gateway library replacement. It provides the self-hosted product surface around gateway routing, governed ChatOps, agents, tools, release delivery, and observability.
LiteLLM is a strong gateway and routing layer. Many teams still need a broader control plane: onboarding flows, tenant administration, ChatOps governance, tool management, agent workflows, and release packaging.
Dekaplane uses LiteLLM under the hood for AI Gateway routing and provider integration while providing the administrative and operational product experience around it.
When a gateway is not enough
Gateway routing matters, but operators also need to manage who can use models, how tools are exposed, what workflows run, and how behavior is traced after execution.
Provider and alias onboarding
Release and license delivery
ChatOps and agent governance
Traceable tool and workflow execution
How Dekaplane relates to LiteLLM
Dekaplane uses LiteLLM as the AI Gateway routing and provider-integration layer. The Dekaplane product experience sits above that layer and focuses on operational administration.
LiteLLM supports gateway routing
Dekaplane stores control-plane configuration
Admins manage onboarding and policy in Dekaplane
Runtime behavior is synchronized behind the scenes
A fair comparison
If your team only needs a routing layer, a gateway project may be enough. If you need a packaged self-hosted control plane with ChatOps, agents, MCP tools, and release access, Dekaplane addresses a broader operating surface.
Dekaplane is currently available as a Free single-tenant self-hosted release with portal-based downloads,
license materials, and installation guidance.